A method and system for round-trip time correction based on reverse path timestamp

Abstract

The application provides a round-trip delay correction method and system based on reverse path timestamp, comprising: collecting data packet sending timestamp, acknowledgement packet sending timestamp and acknowledgement packet receiving timestamp; storing reverse path delay samples dynamically through a sliding window, and calculating a reference delay by selecting a minimum value method or a least square regression method according to the sample quantity; removing reverse path congestion interference in the round-trip delay according to the difference between the current reverse path delay and the reference delay, and obtaining a corrected round-trip delay; and sharing the reference delay of the same source-destination pair under a symmetric topology. The application can accurately separate forward and reverse path congestion, avoid misjudgment of the traditional RTT caused by reverse path congestion, and improve the bandwidth utilization and short flow performance of the congestion control algorithm; through hardware timestamp and dynamic baseline estimation, low-cost deployment without hardware modification is realized, and the application is suitable for high-precision delay optimization of a data center network.

Description

Technical Field

[0001] This invention relates to the field of congestion control algorithm technology, and in particular to a round-trip delay correction method and system based on reverse path timestamps. Background Technology

[0002] As data center networks evolve, the demand for low-latency, high-throughput network performance is increasing. To effectively support emerging applications such as real-time data analytics, AI training, and high-speed storage access (e.g., RDMA, Flash, and NVMe technologies), congestion control mechanisms in data center networks are becoming increasingly critical.

[0003] Currently, mainstream congestion control algorithms (such as Timely and Swift) typically use Round-Trip Time (RTT) as the congestion feedback signal. This mechanism determines network congestion by increasing RTT, thereby dynamically adjusting the sending rate and effectively controlling queue length. It is simple to implement and has good performance. However, it cannot effectively distinguish between congestion on the forward and reverse paths within the RTT. RTT includes both the transmission delay of data packets on the forward path and the return delay of acknowledgment (ACK) packets on the reverse path. When there is congestion on the reverse path, even if there is no congestion on the forward path, the sender may misjudge the network congestion situation, incorrectly reduce the sending rate, and lead to bandwidth waste, resource waste, and network performance degradation.

[0004] Existing technologies generally employ Network Time Protocol (NTP), Precision Time Protocol (PTP), or Data Center Time Protocol (DTP) for clock synchronization to separate forward and reverse path delays. Specifically, while NTP is inexpensive and easy to implement, its synchronization accuracy cannot meet the requirements of microsecond-level latency applications in data centers. Although Precision Time Protocol and DTP offer high clock synchronization accuracy, they rely on dedicated hardware support, resulting in significant network equipment upgrade costs and limiting their practical application scale.

[0005] Furthermore, existing technologies (such as DX) require frequent remeasurement of the baseline delay, resulting in slow convergence when network conditions change or congestion fluctuates, impacting short-flow performance. In addition, these solutions do not consider that ACK packets typically have higher network transmission priority, affecting the accuracy and real-time performance of their baseline delay estimation, making it difficult to meet the high real-time and accurate congestion control requirements of data centers.

[0006] Therefore, there is an urgent need for an RTT correction mechanism that requires no hardware modification, is lightweight, and is easy to deploy on a large scale, in order to address the impact of reverse path congestion on RTT. Summary of the Invention

[0007] In view of this, embodiments of the present invention provide a round-trip delay correction method and system based on reverse path timestamps to eliminate or improve one or more defects existing in the prior art.

[0008] On one hand, the present invention provides a round-trip delay correction method based on reverse path timestamps, the method comprising the following steps:

[0009] Collect the data packet sending timestamp, acknowledgment packet sending timestamp, and acknowledgment packet receiving timestamp; calculate the reverse path transmission delay based on the acknowledgment packet sending timestamp and acknowledgment packet receiving timestamp;

[0010] Maintain a fixed-size sliding window, dynamically store a preset number of reverse path transmission delay samples within a preset time period, and construct a reverse path transmission delay sample set;

[0011] When the amount of data in the reverse path transmission delay sample set is less than a preset value, the minimum value is selected from the reverse path transmission delay samples in the sliding window as the reference delay; when the amount of data in the reverse path transmission delay sample set is not less than the preset value, the least squares regression method is used on the reverse path transmission delay samples in the sliding window to fit the theoretical propagation delay in the queuing-free state, and the theoretical propagation delay is used as the reference delay.

[0012] The round-trip delay is calculated based on the data packet sending timestamp and the acknowledgment packet receiving timestamp; the difference between the current reverse path transmission delay and the reference delay is calculated to obtain the additional queuing delay; the additional queuing delay is removed from the round-trip delay to obtain the corrected round-trip delay.

[0013] In some embodiments of the present invention, the method further includes:

[0014] The data packet sending timestamp, the acknowledgment packet sending timestamp, and the acknowledgment packet receiving timestamp are transmitted through a cross-layer timestamp frame structure;

[0015] The frame structure includes standard protocol fields and extended protocol fields for the link layer, network layer, and transport layer. The extended protocol fields are used to carry timestamp information.

[0016] In some embodiments of the present invention, the method further includes:

[0017] The frame structure reports the timestamp information to user space using single-root I / O virtualization technology or a data plane development kit.

[0018] In some embodiments of the present invention, the reverse path transmission delay is calculated based on the acknowledgment packet sending timestamp and the acknowledgment packet receiving timestamp, and the calculation formula is as follows:

[0019] T b =t4-t3;

[0020] Among them, T b t4 represents the reverse path transmission delay; t3 represents the timestamp of the acknowledgment packet transmission; t4 represents the timestamp of the acknowledgment packet reception.

[0021] In some embodiments of the present invention, a fixed-size sliding window is maintained to dynamically store a preset number of reverse path transmission delay samples within a preset time period, thereby constructing a reverse path transmission delay sample set, which is represented as follows:

[0022]

[0023] Wherein, T represents the reverse path transmission delay sample set; This represents the W-th reverse path transmission delay sample in the sliding window; W is the size of the sliding window.

[0024] In some embodiments of the present invention, when the amount of data in the reverse path transmission delay sample set is less than a preset value, the minimum value is selected from the reverse path transmission delay samples of the sliding window as the reference delay, and the calculation formula is:

[0025]

[0026] in, The reference delay is represented by min{·}, which indicates taking the minimum value.

[0027] In some embodiments of the present invention, when the data volume of the reverse path transmission delay sample set is not less than a preset value, the least squares regression method is used on the reverse path transmission delay samples in the sliding window to fit the theoretical propagation delay under no-queue conditions, and the theoretical propagation delay is used as the reference delay, including:

[0028] Using the confirmation packet sending timestamp as the independent variable and the reverse path transmission delay as the dependent variable, a linear model is fitted. The formula for the linear model is:

[0029]

[0030] The optimal solution to the linear model formula is obtained by using the least squares method. The calculation formula is as follows:

[0031]

[0032] in, This represents the transmission delay sample of the i-th reverse path in the sliding window, where i ∈ W; α represents the clock drift rate; The timestamp of the i-th acknowledgment packet sent in the sliding window is represented by β; β represents the fitting intercept; ε i This represents the measurement error term; n represents the number of currently valid samples in the sliding window;

[0033] The fitted intercept obtained from the solution is used as the reference time delay, and the calculation formula is:

[0034]

[0035] in, This represents the reference delay.

[0036] In some embodiments of the present invention, the round-trip time is calculated based on the data packet transmission timestamp and the acknowledgment packet reception timestamp; the difference between the current reverse path transmission time and the reference time is calculated to obtain the additional queuing time; the additional queuing time is removed from the round-trip time to obtain the corrected round-trip time, including:

[0037] The round-trip time is calculated by subtracting the acknowledgment packet reception time from the data packet transmission timestamp. The calculation formula is as follows:

[0038] RTT = t4 - t1;

[0039] The additional queuing delay is calculated by subtracting the current reverse path transmission delay from the reference delay, using the following formula:

[0040] The additional queuing delay is removed from the round-trip delay to obtain the corrected round-trip delay, calculated as follows:

[0041] RTT corrected =RTT-ΔT b ;

[0042] Where RTT represents the original round-trip time; t4 represents the timestamp of the acknowledgment packet reception; t1 represents the timestamp of the data packet transmission; ΔT b T represents the additional queuing delay; b Indicates the reverse path transmission delay; Represents the reference delay; RTT corrected This indicates the corrected round-trip time delay.

[0043] In some embodiments of the present invention, the method further includes:

[0044] In a symmetric path, connections with the same source and destination addresses share the same baseline latency for the reverse path.

[0045] On the other hand, the present invention also provides a round-trip delay correction system based on reverse path timestamps, wherein the system, when executed, implements the steps of any of the methods mentioned above, and the system includes:

[0046] The timestamp acquisition module is used to collect data packet sending timestamps, acknowledgment packet sending timestamps, and acknowledgment packet receiving timestamps through the network card hardware.

[0047] The baseline delay estimation module is used to maintain a fixed-size sliding window to store reverse path transmission delay samples and dynamically select an estimation strategy based on the number of samples to generate a baseline delay; the estimation strategies include minimum estimation and least squares regression.

[0048] The round-trip delay correction module is used to correct the round-trip delay based on the data packet sending timestamp, the acknowledgment packet sending timestamp, the acknowledgment packet receiving timestamp, and the reference delay.

[0049] This invention provides a method and system for round-trip delay correction based on reverse path timestamps, comprising: collecting data packet transmission timestamps, acknowledgment packet transmission timestamps, and acknowledgment packet reception timestamps; dynamically storing reverse path delay samples through a sliding window, and calculating a baseline delay using either the minimum value method or the least squares regression method based on the sample size; eliminating reverse path congestion interference in the round-trip delay based on the difference between the current reverse path delay and the baseline delay to obtain the corrected round-trip delay; and sharing the baseline delay of the same source-destination pair under a symmetrical topology. This invention can accurately separate forward and reverse path congestion, avoiding misjudgments caused by reverse path congestion in traditional RTT, and improving the bandwidth utilization and short-circuit performance of congestion control algorithms; through hardware timestamps and dynamic baseline estimation, it achieves low-cost deployment without hardware modifications, and is suitable for high-precision delay optimization in data center networks.

[0050] Additional advantages, objects, and features of the invention will be set forth in part in the description which follows, and will also become apparent in part to those skilled in the art upon studying the description, or may be learned by practice of the invention. The objects and other advantages of the invention can be realized and obtained by means of the structures specifically pointed out in the description and drawings.

[0051] Those skilled in the art will understand that the objectives and advantages achievable with the present invention are not limited to those specifically described above, and that the above and other objectives achievable with the present invention will become clearer from the following detailed description. Attached Figure Description

[0052] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this application, are not intended to limit the scope of the invention. In the drawings:

[0053] Figure 1 This is a schematic diagram of the steps of a round-trip delay correction method based on reverse path timestamps in one embodiment of the present invention.

[0054] Figure 2 This is a schematic diagram illustrating the collection of data packet sending and receiving timestamps and acknowledgment packet sending and receiving timestamps between any two communicating parties in one embodiment of the present invention.

[0055] Figure 3 This is a typical architecture diagram of a data center network in one embodiment of the present invention. Detailed Implementation

[0056] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the embodiments and accompanying drawings. Here, the illustrative embodiments and descriptions of this invention are used to explain the invention, but are not intended to limit the invention.

[0057] It should also be noted that, in order to avoid obscuring the invention with unnecessary details, only the structures and / or processing steps closely related to the solution according to the invention are shown in the accompanying drawings, while other details that are not closely related to the invention are omitted.

[0058] It should be emphasized that the term "including / comprises" as used herein refers to the presence of a feature, element, step, or component, but does not exclude the presence or addition of one or more other features, elements, steps, or components.

[0059] It should also be noted that, unless otherwise specified, the term "connection" in this article can refer not only to a direct connection, but also to an indirect connection involving an intermediary.

[0060] In the following description, embodiments of the invention will be illustrated with reference to the accompanying drawings. In the drawings, the same reference numerals represent the same or similar parts, or the same or similar steps.

[0061] It should be emphasized here that the step markers mentioned below are not a limitation on the order of the steps, but should be understood as meaning that the steps can be executed in the order mentioned in the embodiments, or in a different order than in the embodiments, or several steps can be executed simultaneously.

[0062] To address the shortcomings of existing congestion control technologies, such as their inability to effectively distinguish between congestion on the forward and reverse paths in round-trip delay, leading to network status misjudgments, bandwidth and resource waste, and network performance degradation; the high cost and low accuracy of existing time synchronization-based methods, making them unsuitable for dynamic network environments; and the low efficiency of existing baseline delay updates, failing to meet the high real-time and accuracy requirements of data centers, this invention provides a round-trip delay correction method based on reverse path timestamps. Figure 1As shown, the method includes the following steps S101 to S104:

[0063] Step S101: Collect the data packet sending timestamp, acknowledgment packet sending timestamp, and acknowledgment packet receiving timestamp; calculate the reverse path transmission delay based on the acknowledgment packet sending timestamp and acknowledgment packet receiving timestamp.

[0064] Step S102: Maintain a fixed-size sliding window, dynamically store a preset number of reverse path transmission delay samples within a preset time period, and construct a reverse path transmission delay sample set.

[0065] Step S103: When the amount of data in the reverse path transmission delay sample set is less than the preset value, select the minimum value from the reverse path transmission delay samples in the sliding window as the benchmark delay; when the amount of data in the reverse path transmission delay sample set is not less than the preset value, use the least squares regression method on the reverse path transmission delay samples in the sliding window to fit the theoretical propagation delay under no-queue state, and use the theoretical propagation delay as the benchmark delay.

[0066] Step S104: Calculate the round-trip delay based on the data packet sending timestamp and the acknowledgment packet receiving timestamp; calculate the difference between the current reverse path transmission delay and the reference delay to obtain the additional queuing delay; remove the additional queuing delay from the round-trip delay to obtain the corrected round-trip delay.

[0067] In step S101, as Figure 2 As shown, any two communicating parties, such as host A and host B, pass through multiple forwarding devices (such as switches or routers).

[0068] Collect the timestamps recorded when host A sends data packets (data packet sending timestamp), the timestamps recorded when host B receives data packets and returns an acknowledgment packet (ACK) (acknowledgment packet sending timestamp), and the timestamps recorded when host A receives the acknowledgment packet (acknowledgment packet receiving timestamp).

[0069] In some embodiments, Rx / Tx timestamps are acquired via network card hardware (such as the PTP function of IntelNIC), and time synchronization is ensured by calibrating the difference between the network card and the host CPU clock through linear mapping.

[0070] In some embodiments, the packet transmission timestamp, acknowledgment packet transmission timestamp, and acknowledgment packet reception timestamp are transmitted through a cross-layer timestamp frame structure. As shown in Table 1, the frame structure includes link layer, network layer, and transport layer fields for identifying the standard protocol stack, as well as an extended protocol field, which carries the timestamp information.

[0071] Table 1

[0072]

[0073] In some embodiments, the above frame structure can report timestamp information to user space through standard Single Root I / O Virtualization (SR-IOV) technology or Data Plane Development Kit (DPDK) without relying on hardware modifications, and carry latency analysis data through extended protocol fields, which is convenient for subsequent calculation modules to use.

[0074] In some embodiments, the reverse path transmission delay is calculated based on the acknowledgment packet sending timestamp and the acknowledgment packet receiving timestamp, as shown in formula (1):

[0075] T b =t4-t3; (1)

[0076] Among them, T b t4 represents the reverse path transmission delay; t3 represents the acknowledgment packet sending timestamp; t4 represents the acknowledgment packet receiving timestamp.

[0077] In step S102, a fixed-size sliding window is maintained to dynamically store a preset number of reverse path transmission delay samples within a preset time period, thereby constructing a reverse path transmission delay sample set.

[0078] In some embodiments, the reverse path transmission delay sample set is as shown in formula (2):

[0079]

[0080] Where T represents the reverse path transmission delay sample set; This represents the W-th reverse path transmission delay sample within the sliding window; W is the size of the sliding window.

[0081] In step S103, different estimation strategies are selected based on the size of the sample data to obtain the baseline time delay.

[0082] When the amount of data in the reverse path transmission delay sample set is less than the preset value, the minimum value estimation method is adopted, and the minimum value is selected from the reverse path transmission delay samples in the sliding window as the reference delay, as shown in formula (3):

[0083]

[0084] in, This represents the reference delay; min{·} indicates taking the minimum value. This represents the W-th reverse path transmission delay sample in the sliding window.

[0085] If the current new reverse path transmission delay sample is smaller, then update the baseline delay as shown in formula (4):

[0086]

[0087] in, This represents the current new reverse path transmission delay sample.

[0088] Otherwise, add the current new reverse path transmission delay sample to the sliding window while keeping the original baseline delay unchanged.

[0089] In some embodiments, to address estimation bias caused by expired old samples, the time range of the sliding window is set, preferably 2 seconds, and the data within the sliding window is periodically rescanned to ensure the timeliness and robustness of the estimation.

[0090] When the amount of data in the reverse path transmission delay sample set is not less than the preset value, the least squares regression method is used on the reverse path transmission delay samples in the sliding window to fit the theoretical propagation delay under the no-queue state, and the theoretical propagation delay is used as the benchmark delay.

[0091] First, using the confirmation packet sending timestamp as the independent variable and the reverse path transmission delay as the dependent variable, a linear model is fitted (assuming that the baseline delay changes slowly over a short period of time), as shown in formula (5):

[0092]

[0093] The optimal solution of the linear model formula is then obtained by using the least squares method, as shown in formulas (6) and (7):

[0094]

[0095] in, Let represent the transmission delay sample of the i-th reverse path in the sliding window, i∈W; α represents the clock drift rate. ε represents the timestamp of the i-th acknowledgment packet sent in the sliding window; β represents the fitting intercept; ε i This represents the measurement error term; n represents the number of currently valid samples in the sliding window.

[0096] Finally, the obtained fitting intercept β is used as the reference time delay, as shown in formula (8):

[0097]

[0098] in, Indicates the reference delay.

[0099] In summary, by selecting either the minimum estimation method or the least squares regression method to calculate the baseline latency based on the sample data size, this dynamic selection mechanism can maintain low implementation complexity and low resource consumption while taking into account accuracy and stability. It is especially suitable for deployment in data center environments that do not have hardware clock synchronization capabilities.

[0100] Among them, such as Figure 3 The diagram shown illustrates a typical data center network architecture. A data center network is a dedicated communication infrastructure for connecting servers, storage devices, and other computing resources within a data center. It aims to achieve high-speed, low-latency, and highly reliable data transmission to support modern applications such as cloud computing, big data, and AI. The background section mentions that round-trip latency is a key signal for congestion control in data center networks. However, existing technologies cannot distinguish between forward and reverse path congestion, leading to bandwidth waste and performance degradation. The reverse path latency correction scheme provided in this invention accurately separates congestion factors through hardware timestamps and dynamic baseline estimation, directly optimizing the throughput and latency of the data center network.

[0101] In some embodiments, in a data center network, multiple connections (TCP streams / RDMA QP, etc.) between the same source-destination host pairs (e.g., host A and host B) typically follow the same physical path (symmetric topology). If each connection independently measures the reverse path reference delay, it would result in: multiple calculations of the reference delay for the same path, wasting CPU and bandwidth resources; and due to differences in measurement time, the reference delays for each connection might differ, affecting the accuracy of round-trip delay correction. Therefore, in this embodiment, connections with the same source and destination addresses share the same reverse path reference delay.

[0102] Furthermore, a reuse determination mechanism is established, and the reuse reference delay must meet the following conditions:

[0103] Five-tuple consistency: The source IP, destination IP, source port, destination port, and transport protocol (such as TCP / UDP) must be exactly the same.

[0104] Hop count consistency: Verify that the path hop counts of two connections are the same through TTL (Time-To-Live) or explicit path probing (such as traceroute).

[0105] Time validity: The shared baseline latency must be within the validity period (e.g., updated within the last 2 minutes). After the timeout, the baseline latency will be recalculated.

[0106] In step S104, the round-trip time is corrected based on the data packet sending timestamp, the acknowledgment packet sending timestamp, the acknowledgment packet receiving timestamp, and the reference delay.

[0107] First, the original round-trip time is calculated by subtracting the acknowledgment packet reception time from the data packet transmission timestamp, as shown in formula (9):

[0108] RTT = t4 - t1; (9)

[0109] The additional queuing delay is calculated by subtracting the current reverse path transmission delay from the baseline delay calculated in step S103, as shown in formula (10):

[0110]

[0111] Finally, the extra queuing delay is removed from the original round-trip time to obtain the corrected round-trip time, as shown in formula (11): RTT corrected =RTT-ΔT b (11)

[0112] Where RTT represents the original round-trip time; t4 represents the acknowledgment packet reception timestamp; t1 represents the data packet transmission timestamp; ΔT b Indicates additional queuing delay; T b Indicates the reverse path transmission delay; Indicates reference delay; RTT corrected This indicates the corrected round-trip time delay.

[0113] The corrected round-trip time more accurately reflects the congestion situation on the forward path, avoiding misjudgments and unnecessary reductions in transmission rate caused by congestion on the reverse path.

[0114] In some embodiments, the round-trip delay correction method based on reverse path timestamps provided by the present invention can be further applied to the network health assessment system of the financial cloud backbone network, optimizing network performance management through accurate delay measurement. Specific implementation methods are as follows:

[0115] I. Construction of Network Health Indicators:

[0116] Reliability assessment: By using the corrected round-trip time (RTT), true congestion on the forward path is identified, avoiding misjudgments caused by congestion on the reverse path, thereby improving transmission reliability. For example, when the corrected RTT consistently exceeds a threshold, an alarm or path switching is triggered.

[0117] Capacity assessment: Additional queuing delay is calculated based on dynamic baseline latency and real-time reverse path latency, the congestion level of the reverse path is quantified, and the remaining network capacity is comprehensively assessed in conjunction with bandwidth utilization.

[0118] II. The synergistic effect of health status and round-trip delay correction:

[0119] Dynamic policy adjustment: If network health indicates a decline in reliability (e.g., an abnormal increase in corrected round-trip latency), automatically adjust congestion control parameters (e.g., the rate descent gradient of the Swift protocol); if the carrying capacity is insufficient (e.g., an increase in additional queuing latency), trigger traffic scheduling or capacity expansion suggestions.

[0120] Data support: The confirmation packet sending timestamp, confirmation packet receiving timestamp, and calculated baseline delay provided by this invention provide high-precision delay data for the health model, replacing traditional inefficient probe measurements.

[0121] The round-trip delay correction method based on reverse path timestamps provided by this invention is applied to the network health assessment system of the financial cloud backbone network. By correcting the round-trip delay, reverse path interference is eliminated, making the health assessment more objective. At the same time, sharing the benchmark delay of the same source-destination pair can reduce the computational overhead of health assessment.

[0122] Corresponding to the round-trip delay correction method based on reverse path timestamps, the present invention also provides a round-trip delay correction system based on reverse path timestamps. This system, when executed, can implement the steps of the round-trip delay correction method based on reverse path timestamps. The system includes:

[0123] The timestamp acquisition module is used to collect the timestamps of data packet transmission, acknowledgment packet transmission, and acknowledgment packet reception via network card hardware.

[0124] The baseline delay estimation module stores reverse path transmission delay samples by maintaining a fixed-size sliding window and dynamically selecting an estimation strategy based on the number of samples to generate a baseline delay. The estimation strategies include minimum estimation and least squares regression.

[0125] The round-trip delay correction module is used to correct the round-trip delay based on the data packet sending timestamp, acknowledgment packet sending timestamp, acknowledgment packet receiving timestamp, and reference delay.

[0126] Corresponding to the above method, the present invention also provides an electronic device including a computer device, the computer device including a processor and a memory, the memory storing computer instructions, the processor executing the computer instructions stored in the memory, and when the computer instructions are executed by the processor, the electronic device performs the steps of the method as described above.

[0127] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the aforementioned edge computing server deployment method. The computer-readable storage medium can be a tangible storage medium, such as random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, floppy disks, hard disks, removable storage disks, CD-ROMs, or any other form of storage medium known in the art.

[0128] Those skilled in the art will understand that the exemplary components, systems, and methods described in conjunction with the embodiments disclosed herein can be implemented in hardware, software, or a combination of both. Whether implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this invention. When implemented in hardware, it can be, for example, electronic circuits, application-specific integrated circuits (ASICs), appropriate firmware, plug-ins, function cards, etc. When implemented in software, the elements of this invention are programs or code segments used to perform the desired tasks. The programs or code segments can be stored in a machine-readable medium or transmitted over a transmission medium or communication link via data signals carried in a carrier wave.

[0129] It should be clarified that the present invention is not limited to the specific configurations and processes described above and shown in the figures. For the sake of brevity, detailed descriptions of known methods are omitted here. In the above embodiments, several specific steps are described and shown as examples. However, the method process of the present invention is not limited to the specific steps described and shown. Those skilled in the art can make various changes, modifications, and additions, or change the order of steps, after understanding the spirit of the present invention.

[0130] In this invention, features described and / or illustrated for one embodiment may be used in the same or similar manner in one or more other embodiments, and / or combined with or in place of features of other embodiments.

[0131] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, various modifications and variations of the embodiments of the present invention are possible. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A round-trip delay correction method based on reverse path timestamps, characterized in that, The method includes the following steps: collecting data packet sending timestamps, acknowledgment packet sending timestamps, and acknowledgment packet receiving timestamps; calculating the reverse path transmission delay based on the acknowledgment packet sending timestamps and acknowledgment packet receiving timestamps; Maintain a fixed-size sliding window, dynamically store a preset number of reverse path transmission delay samples within a preset time period, and construct a reverse path transmission delay sample set; When the amount of data in the reverse path transmission delay sample set is less than a preset value, the minimum value is selected from the reverse path transmission delay samples in the sliding window as the reference delay; when the amount of data in the reverse path transmission delay sample set is not less than the preset value, the least squares regression method is used on the reverse path transmission delay samples in the sliding window to fit the theoretical propagation delay in the queuing-free state, and the theoretical propagation delay is used as the reference delay. The round-trip delay is calculated based on the data packet sending timestamp and the acknowledgment packet receiving timestamp; the difference between the current reverse path transmission delay and the reference delay is calculated to obtain the additional queuing delay; the additional queuing delay is removed from the round-trip delay to obtain the corrected round-trip delay.

2. The round-trip delay correction method based on reverse path timestamps according to claim 1, characterized in that, The method further includes: The data packet sending timestamp, the acknowledgment packet sending timestamp, and the acknowledgment packet receiving timestamp are transmitted through a cross-layer timestamp frame structure; The frame structure includes standard protocol fields and extended protocol fields for the link layer, network layer, and transport layer. The extended protocol fields are used to carry timestamp information.

3. The round-trip delay correction method based on reverse path timestamps according to claim 2, characterized in that, The method further includes: The frame structure reports the timestamp information to user space using single-root I / O virtualization technology or a data plane development kit.

4. The round-trip delay correction method based on reverse path timestamps according to claim 1, characterized in that, The reverse path transmission delay is calculated based on the timestamp of the acknowledgment packet transmission and the timestamp of the acknowledgment packet reception, using the following formula: T b =t4-t3; Among them, T b t4 represents the reverse path transmission delay; t3 represents the timestamp of the acknowledgment packet transmission; t4 represents the timestamp of the acknowledgment packet reception.

5. The round-trip delay correction method based on reverse path timestamps according to claim 1, characterized in that, Maintain a fixed-size sliding window to dynamically store a preset number of reverse path transmission delay samples within a preset time period, constructing a reverse path transmission delay sample set, which is represented as: Wherein, T represents the reverse path transmission delay sample set; This represents the W-th reverse path transmission delay sample in the sliding window; W is the size of the sliding window.

6. The round-trip delay correction method based on reverse path timestamps according to claim 5, characterized in that, When the amount of data in the reverse path transmission delay sample set is less than a preset value, the minimum value is selected from the reverse path transmission delay samples in the sliding window as the reference delay, and the calculation formula is: in, The reference delay is represented by min{·}, which means taking the minimum value.

7. The round-trip delay correction method based on reverse path timestamps according to claim 5, characterized in that, When the data volume of the reverse path transmission delay sample set is not less than a preset value, the least squares regression method is used on the reverse path transmission delay samples in the sliding window to fit the theoretical propagation delay under no-queue conditions. The theoretical propagation delay is used as the baseline delay, including: Using the confirmation packet sending timestamp as the independent variable and the reverse path transmission delay as the dependent variable, a linear model is fitted. The formula for the linear model is: The optimal solution to the linear model formula is obtained by using the least squares method. The calculation formula is as follows: in, This represents the transmission delay sample of the i-th reverse path in the sliding window, where i ∈ W; α represents the clock drift rate; The timestamp of the i-th acknowledgment packet sent in the sliding window is represented by β; β represents the fitting intercept; ε i This represents the measurement error term; n represents the number of currently valid samples in the sliding window; The fitted intercept obtained from the solution is used as the reference time delay, and the calculation formula is: in, This represents the reference delay.

8. The round-trip delay correction method based on reverse path timestamps according to claim 1, characterized in that, Calculate the round-trip delay based on the data packet sending timestamp and the acknowledgment packet receiving timestamp; Calculate the difference between the current reverse path transmission delay and the reference delay to obtain the additional queuing delay; The additional queuing delay is removed from the round-trip delay to obtain the corrected round-trip delay, including: The round-trip time is calculated by subtracting the acknowledgment packet reception time from the data packet transmission timestamp. The calculation formula is as follows: RTT = t4 - t1; The additional queuing delay is calculated by subtracting the current reverse path transmission delay from the reference delay, using the following formula: The additional queuing delay is removed from the round-trip delay to obtain the corrected round-trip delay, calculated as follows: RTT corrected =RTT-ΔT b ; Where RTT represents the original round-trip time; t4 represents the timestamp of the acknowledgment packet reception; t1 represents the timestamp of the data packet transmission; ΔT b T represents the additional queuing delay; b Indicates the reverse path transmission delay; Represents the reference delay; RTT corrected This indicates the corrected round-trip time delay.

9. The round-trip delay correction method based on reverse path timestamps according to claim 1, characterized in that, The method further includes: In a symmetric path, connections with the same source and destination addresses share the same baseline latency for the reverse path.

10. A round-trip delay correction system based on reverse path timestamps, characterized in that, When the system is executed, it implements the steps of the method as described in any one of claims 1 to 9, and the system comprises: The timestamp acquisition module is used to collect data packet sending timestamps, acknowledgment packet sending timestamps, and acknowledgment packet receiving timestamps through the network card hardware. The baseline delay estimation module is used to maintain a fixed-size sliding window to store reverse path transmission delay samples and dynamically select an estimation strategy based on the number of samples to generate a baseline delay; the estimation strategies include minimum estimation and least squares regression. The round-trip delay correction module is used to correct the round-trip delay based on the data packet sending timestamp, the acknowledgment packet sending timestamp, the acknowledgment packet receiving timestamp, and the reference delay.

Images